Finfish aquaculture, particularly in the oceans, faces a slew of challenges, including disease outbreaks, high mortality rates, water pollution, fish escapes, and feed problems. AI systems are being introduced to tackle these issues, particularly in the U.S., but the solutions are newly emerging and not yet widely adopted.

Some AI models try to predict disease outbreaks or track when nets need repairing. Some are also programmed to act without human input—for example, administering laser energy to kill sea lice, a major problem in the salmon industry. Most, however, focus on improving feed efficiencies, since feed averages 60 percent of a fish farm’s costs.

The models often combine computer vision, which gives cameras the ability to detect and analyze images just as a human would, with machine learning, which enables machines to learn from data and improve over time without further programming.

More than 90 companies are developing AI tools for aquaculture, with the majority headquartered in Norway or the United States, according to new research by ReThink Priorities, a think and do tank. Seventy-six percent of the innovations focus on finfish, especially high-value species like salmon but also trout and tilapia. Most systems are used by large industrial fish farms. That is changing, though, as AI companies develop business models to accommodate smaller-scale operators.

The jury is still out, however, on whether AI applications can help the industry address its most pressing problems. And there are concerns, as in other sectors being transformed by AI, about data quality and privacy, high costs, and the impacts on labor.

A fish farm off the coast of Iceland. (Photo credit: Ed Wingate, Unsplash)

Where Is AI Deployed in Aquaculture?

Thus far, AI adoption worldwide is concentrated among a small number of large companies. Eric Enno Tamm, CEO and co-founder of British Columbia–based seafood software company ThisFish, estimated that in 2024, the seafood industry invested more than $610 million on AI-related initiatives—and that 10 of the world’s largest aquaculture companies had made 86 percent of those investments.

Many of these companies are in Chile or Norway. The latter is the epicenter for this type of AI innovation, with its super-size farms and their data-rich environment attracting AI entrepreneurs.

In the U.S., which has less than 1 percent of the world’s aquaculture, a small number of companies are using AI. Cooke Aquaculture, which produces at least 13,000 metric tons of salmon annually in Maine’s coastal waters, and Atlantic Sapphire, a land-based system in southern Florida that produced 5,096 metric tons of salmon last year, for instance, use AI systems to monitor their stock and optimize feed. Blue Ocean Mariculture, a smaller Hawaiian farm raising kanpachi (Seriola rivoliana), is experimenting with AI to measure fish weight, inventory, and behavior.

Blue Ocean Mariculture also collaborates with the National Oceanographic and Atmospheric Administration (NOAA) on a project using computer vision to monitor and evaluate the behavior of endangered Hawaiian monk seals frequenting the farm. After a baseline study showed where the seals were pulling fish from the pens, the company made equipment changes. This included smaller mesh sizes, a new nursery pen designed for juveniles, and a “mortality trap” with a slide and one-way door that collects and seals off dead fish from predators, Tyler Korte, Blue Ocean’s vice president of marine operations, told Civil Eats.

AI-assisted vision technology captures data points for biomass measurement at a salmon farm. (Photo courtesy of ReelData)

AI Tools for Feeding Fish

Most of the current AI innovation in aquaculture focuses on efficiencies in feeding. Unlike land farmers, fish farmers cannot see their animals to monitor their appetite levels or health status. They estimate how much feed to give thousands to millions of animals by netting out and manually weighing a few hundred fish from each pen or tank every month and working up an average, which they then apply to their total stock.

“There’s a lot of error that goes in with that sampling bias,” said Mathew Zimola, founder of ReelData, an AI company. He added, “Farmers don’t want to touch their fish, because it stresses them out.”

ReelData uses hardware and software systems that enable land-based aquaculture farmers to monitor their fish biomass and automate feeding, without having to scoop them from the water. Its systems deploy underwater cameras equipped with computer vision—a gamechanger for monitoring fish populations on farms and in the wild—to measure and analyze fish in a tank as they swim past.

“Farmers don’t want to touch their fish, because it stresses them out.”

AI tools for optimizing feed are also gaining traction in the shrimp industry, said Dominique Bureau, professor of Animal Nutrition and Aquaculture at the University of Guelph in Canada. These systems use an AI-enabled, underwater microphone, developed by AQ1, that listens for the clicking sound that shrimp make when they bite. Using machine learning, these tools can predict when the animals need to be fed.

“What we’re seeing, based on data, is a lot of overfeeding,” Bureau said. Excess food, he explained, nourishes potentially pathogenic bacteria in the water that produce toxins harmful to shrimp. “Basically you kill your crop.” Automatic feeders coupled with AI microphones are “a great improvement,” he said.

The U.S. produces a fraction of the world’s farmed shrimp. Most farms are very small, though some may use AI tools for measuring shrimp weight and length, such as those developed by Tomota, said Brian Vinci, director of the nonprofit Conservation Fund’s Freshwater Institute.

Imperfect Solutions

AI applications for aquaculture suffer from broader challenges inherent in these technologies: costliness, data quality, availability, privacy issues, and unproven effectiveness. Job loss is also a concern, though it’s too early to tell what impact these emerging systems may have on labor.

Industry insiders agree that few fish farms have robust enough data-management systems for deploying cost-effective AI tools. Uneven data quality can contribute to glitches.

“One of the problems with AI is, it sometimes gives you unpredictable results that can mess up your system,” said Rakesh Ranjan, a research scientist at the Freshwater Institute, who develops AI tools for land-based aquaculture. “There should be a human in the loop who can check and then do the final action.”

Data privacy is also a concern as fish farmers hand over sensitive business and technical data to AI companies, which sometimes anonymize and share their data with feed and pharmaceutical companies and other suppliers to help them hone their products. Regulations are needed to protect data privacy, and farmers need to be clear when they sign agreements with AI companies about how and where their data will be used, Rajan said.

“We’d love to see more peer-reviewed data, as opposed to self-reported data from the companies themselves.”

The costs can be a barrier for many farmers. AI’s high price tags have favored adoption by the biggest players. The AI hydrophone, for example, costs tens of thousands of dollars per pond, Bureau said. However, AI companies are now offering aquaculture tools for a monthly service fee, making them more accessible to a wider range of operations. ReelData’s biomass and feed programs cost a couple thousand dollars a month, Zimola said. “Any farm that’s producing at least 300 metric tons of fish a year can get value from our system.” AI-equipped cameras, which can cost a million dollars to buy, can also now be rented.

And performance claims aren’t yet backed by independent data. ReelData’s feed program increases the growth rate for fish by an average of 10 percent, “turning a 1,000-metric-ton farm into a 1,100 metric-ton-farm,” Zimola said. The program reduces the feed conversion ratio (the amount of feed going into a system versus the amount of food produced) by about 10 percent. Though he’s confident in those numbers, Zimola said, he “isn’t there yet” with publishing the data.

In fact, no independent studies yet exist to evaluate how these emerging AI systems are improving feed operations, fish health, or water quality. “We’d love to see more peer-reviewed data, as opposed to self-reported data from the companies themselves,” said Sophie Williamson, senior researcher in animal welfare at ReThink Priorities.

A Double-Edged Sword?

While AI-enabled tools may solve some industry challenges, they could also lead to larger farms—and the impacts that come with them. “Detecting earlier disease, improving water quality, reducing overfeeding are all positive things for animals,” said Williamson. But there’s a flip side, she said: “We would expect farms may want to stock more aquatic animals.”

Williamson is particularly concerned that farms may increase stocking densities if AI helps them reduce or better monitor disease, and that could have other animal welfare impacts.

Alternatively, as companies increase efficiencies and reduce costs, they may expand the size of their farms. That is likely true, according to Tony Chen, founder of Manolin Aqua, an AI service that uses machine learning to predict fish health. “When you talk to farmers, they all want to make more money [and] bring more fish to the market, but . . . the piece that created uncertainty in their businesses is fish health.”

The MARA Act, which supports deep-ocean aquaculture research, is moving through Congress. If it passes, the legislation could lead to larger fish farms in U.S. waters, and potentially a wider adoption of AI systems within the industry.

For Vinci, AI tools might help clean up some of the salmon industry’s problems. “I do think we’ll see improvements in things like identifying fish behaviors that show stress, or showing ulcers on fish or lice on fish, and having ways to quickly address that before it gets out of control,” he said.

The post Will AI Be a Net Positive For Aquaculture? appeared first on Civil Eats.

From Civil Eats via This RSS Feed.